• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 47 subscribers
  • Views 763 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Original question:

INA241A-Q1: Using a negative power rail (1.2V) to the GND of the chip.

INA241A: unstable of INA241A5

Manu Chang
Expert 3300 points

Part Number: INA241A
Other Parts Discussed in Thread: LM324

Hi Team,

My customer is using INA241A5 with 200V/V gain and a 0.4mohm shunt current to sampling the maximum 50A current. The supply of VSS and GND is 12V and -7V for sensing the bidirectional current.

To maintain the high CMRR, they add capacitors on IN+ and IN-, because they observed that when the common mode voltage changes from 3V to 2.9V there is a 100mA error of INA241. So they add a 4.7uF capacitor on IN- and IN+. However, the problem of the response time then exists. When there is a 50A through the shunt for 2ms, the output voltage is quite unstable. It seems stable when the time raise to 10ms (50A lasts for 10ms). The output signal can be read as a stable signal. Based on the trade off between the CMRR and response/settle time, customers have some questions:

1. Are there any methods to achieve the better CMRR and the faster settle time ? Currently it seems customers cannot find a good way to balance this, hope that you can give some advices.  

2. For the 12V and -7V supply for INA241, could this asymmetric supply voltage be a problem?

Pls give some suggestion on this, thanks!

Manu

  • 0
    TI__Mastermind 34840 points

    Hello Manu,

    Please share a schematic with annotation of the voltages as I do not see why you are having issues or why capacitors are needed.  Please make sure to add Vref and output conduins or loads.

    Javier

  • 0 in reply to Javier Contreras4
    TI__Expert 3300 points

    Hi Javier,

    According to policy their schematic cannot be sent out, so pls check the drawn schematic instead. the output is connected to a follower amplifier and then goes to the input of ADC. They add C1 and C2 at IN- and IN+ pin because the data read by ADC is not stable and decreasing. Adding 4.7uF C1 and C2 makes the data stable but the response time is at least 10ms. 

    The INA241 is used to measure the current through battery cell, the voltage of cell, namely common mode voltage, is around 5V and maximum 8V.

    Any suggestion to achieve the better CMRR (stable read data) and faster settle time? Pls give some insights, thanks!

    BR,

    Manu 

  • 0 in reply to Manu Chang
    TI__Mastermind 34840 points

    Helllo Manu,

    The INA241 cannot drive a capacitance that high.  According to datasheet you will have issues driving the 10nF capacitor as it is rated for 1nF(see image below).  If you place the 1kΩ before the capacitor that will solve that issue.  The issue with adding common mode capacitors is the mismatch from these devices will cause a larger VCM transient difference and may propagate to the output.  If this is the way the circuit is done I think the reason for unstable measurement is the 10nF at the output.  Please verify this by removing.

    Let me know if this does not solve the issue.

    Regards,

    Javier Contreras

  • 0 in reply to Javier Contreras4
    TI__Expert 3300 points

    Hi Javier,

    Thanks for your response. I think I missed drawing the 1kohm resistor before the 10nF capacitor in schematic, sorry for that. I double checked with customers, in their evaluation, at the output there is a 1kohm resistor before 10nF capacitor and another 1kohm after 10nF capacitor. 

    In your response, you mentioned that 1kohm before the 10nF can solve this. Actually customers have added 1kohm before but it is still have some stable issue. So do you think that the unstable still should come from the 10nF at output and I should remove this and 1kohm resistors? Also do customers need to remove the C1 and C2 as well?

    BR,

    Manu

  • 0 in reply to Manu Chang
    TI__Mastermind 34840 points

    Hello Manu,

    I think that there is another issue where there could be the wrong voltages from another pin that is not expected.  There should not be difference when the voltage goes from 2.9V to 3.0V.  From the INA241 standpoint if the schematic is as above with the resistor of 1kΩ before the capacitor I do not see any issues.  Therefore I will as more questions.

    Also is the LM324 fast enough to drive a SAR ADC? https://www.ti.com/video/series/precision-labs/ti-precision-labs-analog-to-digital-converters-adcs.html

      Have they placed a scope at different points on the circuit to see the signal?  By unstable is it just the ADC output or is it at the output of the INA241.  Does the ADC support negative voltages?  Have they measured the voltage at the output of the INA241 before the ADC?  Do they have an expected measurement vs the measurement they are getting?